This research addresses two fundamental but poorly understood areas of the biology and biochemistry of microtubules and microtubule proteins: the physiological regulation of microtubule assembly, and the existence and consequence of direct interaction between membranes and the principal protein subunit of microtubules, tubulin. In previous studies of microtubule assembly/disassembly, we showed that an increase in pH raises the critical concentration for assembly and increases the drug sensitivity of microtubule in vitro. Our goal now is to determine the role of pH in in vivo microtubule regulation. Using new spectroscopic methods, we will establish the relationship of microtubule disassembly and pH under physiological conditions in the J774 mouse macrophage, determine the role of Na[unreadable]+[unreadable]-H[unreadable]+[unreadable] exchange in alkalinization, and extend the analysis to a neuroblastoma line in which spontaneous microtubule disassembly/assembly can be correlated with pH in single cells. We have employed a strategy for isolation of membrane tubulin from bovine brain synaptosomes that has revealed hydrophobic tubulins that are chemically distinct from soluble tubulins but which cross-react with a monoclonal tubulin antibody and copolymerize with soluble bovine brain tubulin. The procedure has also uncovered several potential hydrophobic microtubule-associated proteins. Briefly, synaptosomal membranes are removed from a sucrose gradient and extracted with 1% Triton X-100 in phosphate-glutamate buffer. The extract is centrifuged at 10[unreadable]5[unreadable]g for 60 min and the pellet discarded. GTP-Mg is added and the supernatant incubated in the presence of 5 micromolar Taxol for 30 min. This mixture is then recentrifuged at 10[unreadable]5[unreadable]g for 30 min through a 15% sucrose cushion. The resulting pellet is then extracted with Triton X-114 and the hydrophobic proteins partitioned according to Bordier. On SDS gels, roughly 50% of the dimeric approximately 55 kilodalton species is found in the hydrophobic extract. This protein cross-reacts wih a monoclonal tubulin antibody on immunoblot but shows distinct differences in peptide map after N-chlorosuccinimide digestion. At least two higher and one lower molecular weight proteins are uniquely present in the extract. These putative MAPs as well as the hydrophobic tubulin described suggest new ways in which membrane functions may be affected by microtubule proteins. (L)